Getting different hash digest values for the same string when using hash functions in a list

Question

I seem to be getting different digest values for the same word in my program. I am not sure this is because I am keeping the hash functions in a list (So I can add to the list)

When I use direct hash functions the hash digest is the same for the same word. It is different when I use the hashes from inside a list. What am I doing wrong ?

What is working

import hashlib

bloom_len = 100

def bytes_to_int(hash_value):
    return int.from_bytes(hash_value, byteorder='big')  #big-endiang format

def bloom_index(hashint):
    return hashint % bloom_len


def hashIt(word):

    m1 = hashlib.md5()
    m2 = hashlib.sha1()
    m3 = hashlib.sha256()

    m4 = hashlib.sha3_512()
    m5 = hashlib.blake2s()

    m1.update(word)
    m2.update(word)
    m3.update(word)

    m4.update(word)
    m5.update(word)


    hash_values = [m1.digest(), m2.digest(), m3.digest(), m4.digest(), m5.digest()]
    hashints = list(map(bytes_to_int, hash_values))
    indices = list(map(bloom_index, hashints))

    print(indices)


inputWord = 'sent'
word = inputWord.encode('utf-8')
hashIt(word)

inputWord = 'blue'
word = inputWord.encode('utf-8')
hashIt(word)

inputWord = 'sent'
word = inputWord.encode('utf-8')
hashIt(word)

What is NOT working

import hashlib


class BloomFilter():
    def __init__(self, length = 100):

        self.bloomFilterLen = length
        self.bloomFilterArray = [0] * self.bloomFilterLen

        m1 = hashlib.md5()
        m2 = hashlib.sha3_512()
        m3 = hashlib.blake2s()        

        self.hashes = [m1, m2, m3]


    def encode(self, inputWord):
        encoded_word = inputWord.encode('utf-8')
        return encoded_word

    def bytes_to_int(self, hash_value):

        return int.from_bytes(hash_value, byteorder='big')  

    def bloom_index(self, hashint):

        return hashint % self.bloomFilterLen    

    def getIndices(self, inputWord):

        word = self.encode(inputWord)

        print(word)
        hashDigests = []

        for hashFunction in self.hashes:
            hashFunction.update(word)
            print('hashFunction ', hashFunction , '\n')
            print('hashDigest ', hashFunction.digest() , '\n')

            hashDigests.append(hashFunction.digest())


        hashInts = [self.bytes_to_int(h) for h in hashDigests]    
        #print('hashInts ', hashInts)

        bloomFilterIndices = [self.bloom_index(hInt) for hInt in hashInts]
        return bloomFilterIndices

    def insert(self, inputWord):

        bloomFilterIndices = self.getIndices(inputWord)

        for index in bloomFilterIndices:
            self.bloomFilterArray[index] = 1

        print(bloomFilterIndices)


    def lookup(self, inputWord):

        bloomFilterIndices = self.getIndices(inputWord)
        print('Inside lookup')
        print(bloomFilterIndices)

        for idx in bloomFilterIndices:
            print('idx value ', idx)
            print('self.bloomFilterArray[idx] value ', self.bloomFilterArray[idx])
            if self.bloomFilterArray[idx] == 0:
                # Indicates word not present in the bloom filter
                return False

        return True            


if __name__ == '__main__':

     word = 'sent'
     bloomFilter = BloomFilter()
     bloomFilter.insert(word)


     print(bloomFilter.lookup(word))

From the first program - I consistently get same integer indices

• Indices for 'sent'

[61, 82, 5, 53, 87]

• Indices for 'blue'

[95, 25, 24, 69, 85]

• Indices for 'sent'

[61, 82, 5, 53, 87]

For the non-working program the integer indices are different and when I printed out the hash digest is different

• Indices for 'sent' - first time via add

[61, 53, 87]

HashDigest from MD5 for 'sent'

hashDigest b'x\x91\x83\xb7\xe9\x86F\xc1\x1d_\x05D\xc8\xf3\xc4\xc9'

• Indices for 'sent' - second time via lookup

[70, 89, 8]

HashDigest from MD5 for 'sent'

hashDigest b'\x95\x17bC\x17\x80\xb5\x9d]x\xca$\xda\x89\x06\x16'

Answer 1

So I changed the code in __init __

From

m1 = hashlib.md5()
m2 = hashlib.sha3_512()
m3 = hashlib.blake2s()        

self.hashes = [m1, m2, m3]

To

self.hashes = ['md5', 'sha3_512', 'blake2s']

And then inside the for loop in method getIndices()

Changed from

  for hashFunction in self.hashes:
        hashFunction.update(word)

To

for hashName in self.hashes:
    hashFunction = hashlib.new(hashName)
    hashFunction.update(word)

Works now !